Hi Ray, Here is the description of the different latencies under filestore perf counters.
Journal_latency : ---------------------- This is the latency of putting the ops in journal. Write is acknowledged after that (well a bit after that, there is one context switch after this). commitcycle_latency: -------------------------- Filestore backend while carrying out transaction, do a buffered write. In a separate thread it does call syncfs() to persist the data to the disk and update the persistent commit number in a separate file. This thread runs by default 5 sec interval. This latency measures the time taken to carry out this job after the timer expires i.e the actual persisting cycle. apply_latency: ---------------- This is the entire latency till the transaction finishes i.e journal write + transaction time. It will do a buffer write here. queue_transaction_latency_avg: ---------------------------------------- This is the latency of putting the op in the journal queue. This will give you an idea how much throttling is going on at the first place. This depends on the following two parameters if you are using XFS. filestore_queue_max_ops filestore_queue_max_bytes All the latency numbers are represented by avgcount(number of ops within this range) and the sum (which is total latency in second). Sum/avgcount will give you an idea the latency per op. Hope this is helpful, Thanks & Regards Somnath From: ceph-users [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of Ray Sun Sent: Sunday, July 05, 2015 7:28 AM To: ceph-users@lists.ceph.com Subject: [ceph-users] Meanning of ceph perf dump Cephers, Is there any documents or code definition to explain ceph perf dump? I am a little confusing about the output, for example, under filestore, there's journal_latency and apply_latency and each of them has avgcount and sum. I am not quite sure what's the unit and meaning of the numbers? How can I use these numbers to tuning my ceph cluster. Thanks a lot. "filestore": { "journal_queue_max_ops": 300, "journal_queue_ops": 0, "journal_ops": 35893, "journal_queue_max_bytes": 33554432, "journal_queue_bytes": 0, "journal_bytes": 20579009432, "journal_latency": { "avgcount": 35893, "sum": 1213.560761279 }, "journal_wr": 34228, "journal_wr_bytes": { "avgcount": 34228, "sum": 20657713152 }, "journal_full": 0, "committing": 0, "commitcycle": 3207, "commitcycle_interval": { "avgcount": 3207, "sum": 16157.379852152 }, "commitcycle_latency": { "avgcount": 3207, "sum": 121.892109010 }, "op_queue_max_ops": 50, "op_queue_ops": 0, "ops": 35893, "op_queue_max_bytes": 104857600, "op_queue_bytes": 0, "bytes": 20578506930, "apply_latency": { "avgcount": 35893, "sum": 1327.974596287 }, "queue_transaction_latency_avg": { "avgcount": 35893, "sum": 0.025993727 } }, Best Regards -- Ray ________________________________ PLEASE NOTE: The information contained in this electronic mail message is intended only for the use of the designated recipient(s) named above. If the reader of this message is not the intended recipient, you are hereby notified that you have received this message in error and that any review, dissemination, distribution, or copying of this message is strictly prohibited. If you have received this communication in error, please notify the sender by telephone or e-mail (as shown above) immediately and destroy any and all copies of this message in your possession (whether hard copies or electronically stored copies).
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
